keyword
denotational uncertainty
Denotational uncertainty refers to the ambiguity or indeterminacy regarding the precise meaning, intent, or formal interpretation of an input expression, such as a natural language question or prompt. In computational linguistics and machine learning, this form of uncertainty occurs when an input is underspecified, lacks adequate background context, or naturally supports multiple distinct interpretations. It is distinct from epistemic uncertainty, which arises when a system lacks the factual knowledge required to answer a clearly stated query; under denotational uncertainty, several valid answers may exist solely because the initial query maps to multiple possible concepts or references. Resolving or calibrating for denotational uncertainty typically involves recognizing linguistic ambiguity, generating explicit disambiguations, or evaluating distributions over potential interpretations.
1 item
